Difference between revisions of "Mapping Networks on Reconfigurable Binary Engine Accelerator"
From iis-projects
(→Links) |
|||
(One intermediate revision by one other user not shown) | |||
Line 8: | Line 8: | ||
===Status: Available === | ===Status: Available === | ||
: Looking for 1-2 Semester/Master students | : Looking for 1-2 Semester/Master students | ||
− | : Contact: [[:User:Paulin| Gianna Paulin]] Thorir Ingolfsson | + | : Contact: [[:User:Paulin| Gianna Paulin]] [[:User:Thoriri| Thorir Mar Ingolfsson]] |
===Prerequisites=== | ===Prerequisites=== | ||
Line 58: | Line 58: | ||
[[Category:Deep Learning Projects]] | [[Category:Deep Learning Projects]] | ||
[[Category:Available]] | [[Category:Available]] | ||
+ | [[Category:Digital]] | ||
[[Category:2020]] | [[Category:2020]] | ||
[[Category:Paulin]] | [[Category:Paulin]] |
Revision as of 18:12, 16 November 2020
Contents
Short Description
We have recently designed an accelerator called Reconfigurable Binary Engine (RBE). The RBE architecture uses these two innovations to emulate quantized NNs by choosing the binary weights to correspond to each bit of the quantized weights. One quantized NN can therefore be emulated by a superposition of power-of-2 weighted Q) × Q+ binary NN, whereas Q+ corresponds to the quantization level of the weights and Q) quantization level of the activations. We call this concept from now on Binary Based Quantization (BBQ) which allows the RBE to perform convolutions with configurable arithmetic precisions in a flexible and power-scalable way. In this project we make use of our in-house developed frameworks NEMO and DORY to map networks onto the RBE accelerator and evaluate its performance and energy-efficiency for real networks.
Status: Available
- Looking for 1-2 Semester/Master students
- Contact: Gianna Paulin Thorir Mar Ingolfsson
Prerequisites
- VLSI I
- C coding
- python coding (optimal: Pytorch)
Character
- 20% Theory
- 20% HW understanding
- 40% ML Tools: Nemo, Dory, Pytorch
- 20% Embedded C programming